The Mechanism of Inhibition of the Ca2+-ATPase by Mastoparan

The amphiphilic peptide mastoparan, isolated from wasp venom, is a potent inhibitor of the sarcoplasmic reticulum Ca2+-ATPase. At pH 7.2, ATPase activity is inhibited with an inhibitory constant (K i ) of 1 ± 0.13 μm. Mastoparan shifts the E2-E1 equilibrium toward E1 and may affect the regulatory ATP binding site. The peptide also decreases the affinity of the ATPase for Ca2+ and abolishes the cooperativity of Ca2+ binding. In the presence of mastoparan, the two Ca2+ ions bind independently of one another. Our results appear to support the model that describes the relationship between the two Ca2+ binding sites as “side-by-side,” because this model allows the possibility of independent Ca2+ entry to the two sites. Mastoparan shifts the steady-state equilibrium between E1′Ca2 and E1′Ca2·P toward E1′Ca2·P, by possibly affecting the conformational change that follows ATP binding. The peptide also causes a reduction in the levels of phosphoenzyme formed from [32P]Pi. Some analogues of mastoparan were also tested and were found to cause inhibition of the Ca2+-ATPase in the range of 2–4 μm. The inhibitory action of mastoparan and its analogues appears dependent on their ability to form α-helices in membranes.

The sarco(endo)plasmic reticulum Ca 2ϩ -ATPase transports Ca 2ϩ from the cytosol to the lumen of the sarcoplasmic reticulum (SR) 1 /endoplasmic reticulum (ER). The mechanism of this Ca 2ϩ -ATPase is usually discussed in terms of the model proposed by de Meis and Vianna (1). The model postulates two major conformational states of the enzyme, E1 and E2. These two states differ in that the affinity for Ca 2ϩ is high in the E1 conformation but low in the E2 conformation, and the Ca 2ϩbinding sites are exposed on the cytoplasmic side of the SR in E1 but exposed to the lumen in E2.
It has been demonstrated that the binding of Ca 2ϩ to the ATPase is both sequential and co-operative (2). This suggests that binding of the first Ca 2ϩ ion is followed by a slow conformational change (E1Ca 3 E1ЈCa), which allows binding of the second Ca 2ϩ ion. The second site is only formed on transition to E1ЈCa (2).
The Ca 2ϩ -ATPase belongs to a family of enzymes known as the P 2 -type ATPases (3). Several peptide toxins have been shown to inhibit the action of these enzymes. Both myotoxin a, from rattle snake venom, and melittin, isolated from bee venom, are basic peptides that inhibit the SR Ca 2ϩ -ATPase (4 -6). Melittin is also a potent inhibitor of the H ϩ /K ϩ -ATPase and the Na ϩ /K ϩ -ATPase (7)(8)(9)(10). Mastoparan (MP) is an amphiphilic tetradecapeptide isolated from wasp venom (11). It is known to possess a variety of biological activities including mast cell degranulation, mobilization of Ca 2ϩ from cerebellar microsomes and sarcoplasmic reticulum, activation of the ryanodine receptor and modulation of various enzymes, for example the Na ϩ /K ϩ -ATPase of rat brain (12)(13)(14).
In aqueous solutions, mastoparan forms a random structure, however, in a lipid environment, the peptide adopts an amphiphilic ␣-helical structure, which is thought to be crucial for its interaction with biological membranes (15). In a previous study (13), we showed that mastoparan and a number of closely related analogues inhibit the SR Ca 2ϩ -ATPase. Here we elucidate the mechanism of this inhibition. SR and the purified Ca 2ϩ -ATPase were prepared from rabbit fasttwitch skeletal muscle as described by Michelangeli and Munkonge (16). Ca 2ϩ -ATPase activities were determined using the coupled enzyme method described by Michelangeli and Munkonge (16) and monitored in a buffer containing 40 mM Hepes/KOH (pH 7.2), 1 mM EGTA, 5 mM MgSO 4 , 2 mM ATP, 0.42 mM phosphoenolpyruvate, 0.15 mM NADH, 7.5 IU of pyruvate kinase, and 18 IU of lactate dehydrogenase. Ca 2ϩ -ATPase (15 g) was incubated for 10 min at 37°C in 2.5 ml of assay buffer. ATPase activity was initiated by the addition of 1 mM CaCl 2 to give a free Ca 2ϩ concentration of 6.5 M.
Fluorescence measurements were performed at 25°C using a Perkin-Elmer LS-50B fluorimeter. Measurements of NBD fluorescence were made at excitation and emission wavelengths of 430 and 510 nm, respectively, in a buffer containing either 150 mM Mops/Tris, 0.3 mM EGTA, 100 mM choline chloride at pH 7.2, or 150 mM Mes/Tris, 0.3 mM EGTA, 100 mM choline chloride at pH 6.0. Tryptophan fluorescence was monitored by exciting at 295 nm and measuring the emission at 330 nm. These measurements were made in a buffer containing 20 mM Hepes/Tris, 100 mM KCl, 5 mM MgSO 4 , 100 M Ca 2ϩ at pH 7.2.
Rapid kinetic fluorescence measurements were performed using a stopped-flow spectrofluorimeter (Applied Photophysics, Model SX17 MV). The sample handling unit possesses two syringes, A and B (drive ratio 10:1), which are driven by a pneumatic ram. Tryptophan fluorescence was monitored (at 25°C) by exciting the sample at 280 nm and measuring the emission above 320 nm using a cut off filter. Filters were left to dry, after which 8 ml of scintillant was added. The filters were then counted for both 3 H and 45 Ca 2ϩ . The amount of [ 3 H]glucose trapped on each filter was used to calculate the wetting volume for the filter, and the amount of Ca 2ϩ trapped in this volume was subtracted from the total Ca 2ϩ bound to the filter to give that bound to the ATPase. A correction was also applied for nonspecific binding of Ca 2ϩ to the filter.
Equilibrium levels of phosphorylation of the ATPase by [ 32 P]P i were measured in 150 mM Mes/Tris (pH 6.2), containing 5 mM EGTA, 10 mM MgSO 4 , and 1 mM P i (10 Ci/mol), at 25°C and a protein concentration of 0.9 mg/ml. Samples were incubated for 20 s and then quenched with 10% trichloroacetic acid, 0.2 M H 3 PO 4 . The precipitate was collected by rapid filtration through Whatman GF/C filters, washed with 30 ml of 12% trichloroacetic acid, 0.2 M H 3 PO 4 , and then counted.
Steady-state levels of phosphorylation of the ATPase by [␥-32 P]ATP were carried out in a similar manner as above. Experiments were carried out at 25°C in 20 mM Hepes/Tris (pH 7.2) containing 100 mM KCl, 5 mM MgSO 4 , 100 M CaCl 2 , and 0.075 mg/ml ATPase. Two stocks of labeled ATP were made up to cover the range of ATP concentrations up to 100 M, with specific activities of 10 and 100 Ci/mol. The reaction was initiated by addition of [␥-32 P]ATP and quenched as described above after 10 s. The samples were then filtered, washed, and counted.
Dual wavelength spectrophotometry was performed on a Shimadzu UV-3000 dual wavelength-recording spectrophotometer. Experiments were carried out at 25°C in 20% (w/v) sucrose, 50 mM Mops/KOH (pH 7) containing 1 mM CaCl 2 and 0.8 mg/ml ATPase. Titration of the Ca 2ϩ -ATPase with trinitrophenyl adenosine diphosphate (TNP-ADP) was then monitored by recording the absorbance difference at 422 nm and 390 nm, as described by Coll and Murphy (20). Fig. 1A shows the effect of mastoparan on purified, fully uncoupled, Ca 2ϩ -ATPase activity. The inhibitory constant is determined to be 1 Ϯ 0.13 M mastoparan. In sealed SR vesicles, ATPase activity is low due to high Ca 2ϩ concentrations in the vesicle lumen, and the inhibition can be relieved by addition of the Ca 2ϩ ionophore A23187. Effects of mastoparan on ATPase activity of SR vesicles in the presence of A23187 are very similar to those determined for the purified ATPase (data not shown).

Ca 2ϩ -ATPase Inhibition-
However, in the absence of A23187, low concentrations of mastoparan (Ͻ1 M) actually increase the activity of the ATPase. Thus mastoparan increases the permeability of the SR membrane to Ca 2ϩ , a phenomenon previously reported by Longland et al. (13). At higher concentrations of mastoparan (Ն1 M), ATPase activity in sealed SR vesicles decreases due to inhibition of the pump. Fig. 1B shows the effect of mastoparan on the Ca 2ϩ dependence of ATPase activity. A bell-shaped curve was obtained both in the presence and absence of mastoparan. The K m value for the high-affinity (activatory) Ca 2ϩ sites was increased from 0.16 M in the absence of peptide to 0.27 M in the presence of mastoparan. In contrast, the K m value for the lower affinity (inhibitory) sites (0.2 mM) was not significantly effected by the inhibitor. In addition, maximum ATPase activities were observed at similar free Ca 2ϩ concentrations in both the presence and absence of mastoparan (i.e. 6.5 M).
The effect of mastoparan on the dependence of ATPase activity on the concentration of ATP is shown in Fig. 1C. The data are fitted to a modified form of the Michaelis-Menten equation, assuming that ATP interacts at 2 sites: a high-affinity (catalytic) site and a low affinity (regulatory) site (21).
Addition of mastoparan had little effect on ATPase activity at low ATP concentrations but considerably inhibited the pump at higher ATP concentrations. In the absence of peptide, the data could be fitted assuming K m and V max values for the catalytic site of 0.85 M and 1.68 IU/mg, respectively, and K m and V max values for the regulatory site of 0.15 mM and 4.34 IU/mg, respectively. In the presence of mastoparan, the data could be fitted assuming the same values for the catalytic site and the same K m value for the regulatory site. The V max value for the regulatory site, however, decreased to 1.24 IU/mg. Fig. 1D shows the effect of mastoparan on the Mg 2ϩ dependence of ATPase activity. In the absence of mastoparan, ATPase activity decreases with increasing concentrations of Mg 2ϩ . At low concentrations of Mg 2ϩ , the presence of mastoparan results in strong inhibition of the ATPase. As the Mg 2ϩ concentration is increased from 2-10 mM, stimulation of ATPase activity is then observed followed by inhibition.
NBD and FITC Fluorescence-It has been shown that the E2-E1 equilibrium for the ATPase can be monitored by changes in the fluorescence intensity of the ATPase labeled with NBD (22). The fluorescence intensity of the labeled ATPase is higher in the E1 conformation than in the E2 conformation (22).
Addition of mastoparan to NBD-labeled SR at pH 7.2 ( Fig. 2) results in an increase in fluorescence intensity with an apparent K d value of 1.8 M. The maximal fluorescence increase observed is 17-18%. Therefore mastoparan shifts the E2-E1 equilibrium toward E1.
It has been suggested that the E2-E1 equilibrium is pH-dependent with low pH favoring the E2 form (23). Consequently pH can also be used to trigger the E2-E1 transition. The addition of mastoparan to NBD-labeled SR at pH 6.0 (Fig. 2) also causes an increase in fluorescence intensity, however, with a higher apparent K d (14.1 M) and a maximal fluorescence increase of Ϸ31%.
The E2-E1 equilibrium can also be studied by monitoring changes in the fluorescence intensity of FITC-labeled ATPase. In contrast to NBD-labeled ATPase, addition of Ca 2ϩ to FITClabeled ATPase results in a decrease in fluorescence, which has been attributed to the E1 conformational state (23). At pH 7.0, where the E1/E2 ratio has been determined to be 0.5 (23), calcium induces a decrease in fluorescence of 6%, whereas in the presence of 20 M mastoparan, this decrease is reduced to 3%. This is consistent with mastoparan having shifted the E2-E1 equilibrium toward the E1 conformation. Furthermore, addition of 20 M mastoparan to FITC-labeled ATPase in the absence of calcium also caused a 3% decrease in fluorescence.
Calcium Binding and Dissociation- Table I shows the level of 45 Ca 2ϩ bound to the ATPase at a free Ca 2ϩ concentration of 50 M, a concentration at which both high-affinity Ca 2ϩ binding sites should be fully saturated. This level is unchanged in the presence of 30 M mastoparan (added either before or after the labeled Ca 2ϩ ), demonstrating that mastoparan does not effect the stoichiometry of Ca 2ϩ binding. Levels of Ca 2ϩ binding to the native ATPase are higher than expected as a result of nonspecific binding of Ca 2ϩ to the ATPase and associated lipids (19). Fig. 3 shows Ca 2ϩ binding to the ATPase as a function of Ca 2ϩ concentration. It clearly shows that mastoparan reduces the affinity of Ca 2ϩ binding to the ATPase, increasing the K d from 0.6 to 3.7 M. In the absence of mastoparan, binding of Ca 2ϩ to the ATPase is cooperative, as expected, with a Hill coefficient of 1.60. In the presence of mastoparan, this cooperativity is no longer observed, with the Hill coefficient being reduced to 0.9.
Calcium binding and dissociation can also be studied through changes in the tryptophan fluorescence of the ATPase. On addition of Ca 2ϩ to the ATPase, there is an increase in tryptophan fluorescence that has been attributed to the E1Ca-E1ЈCa transition, with the E2, E1, and E1Ca forms having relatively low tryptophan intensities and the E1ЈCa and E1ЈCa 2 forms having higher fluorescence intensities (24). Fig. 4 shows that the addition of 20 M mastoparan to the ATPase causes a shift in the Ca 2ϩ concentration dependence of this transition to higher Ca 2ϩ concentrations. The apparent K d value is increased from 1.4 to 25 M, suggesting once again that mastoparan may decrease the affinity of the ATPase for Ca 2ϩ . The presence of the peptide also halves the maximum change in the fluorescence intensity observed.
It has been shown in stopped-flow experiments with the Ca 2ϩ -ATPase that both Ca 2ϩ binding and dissociation are biphasic in nature (25). Fig. 5 shows the binding (A) and dissociation (B) of calcium to and from the ATPase in the absence and presence of 30 M mastoparan. The kinetic parameters obtained from these experiments are given in Table II. In the absence of peptide, the data for both Ca 2ϩ binding and dissociation can be fitted to the following biexponential equation (fitting these data to a monoexponential equation resulted in much larger 2 values, see Table II) where ⌬F equals fluorescence change, A 1 , A 2 , k 1 , and k 2 are the amplitudes and rate constants for the fast and slow phases of Ca 2ϩ binding/dissociation, respectively, and t is time (s).  In the presence of mastoparan, the data for both Ca 2ϩ binding and dissociation can be fitted equally well to the following monoexponential equation.

Phosphorylation of the ATPase by [␥-32 P]ATP and [ 32 P]P i -Steady-state levels of phosphorylation can be studied under conditions of low Ca 2ϩ
where the ATPase will be in a continuous state of turnover. At 25°C, mastoparan increases the steady-state levels of phosphoenzyme formation over the range of ATP concentrations used (Fig. 6). In the absence of mastoparan, apparent K d and EP max values of 46 M and 9.1 nmol EP/mg ATPase, respectively, are obtained, whereas in the presence of mastoparan, these values are reduced to 9 M and 9.1 nmol EP/mg. These results suggest that mastoparan has either increased the affinity of the ATPase for ATP and/or that the steady-state equilibrium between E1ЈCa 2 and E1ЈCa 2 ⅐P has been shifted toward E1ЈCa 2 ⅐P.
TNP-ADP is known to be a competitve inhibitor of the SR Ca 2ϩ -ATPase (26,27). It binds very tightly to the active site of the enzyme. Consequently, if the nucleotide binding site of the ATPase is covalently blocked with FITC, TNP-ADP is unable to bind. As described under "Materials and Methods," dual wavelength spectrophotometry can be used to monitor the absorbance difference at 422 and 390 nm on titration of Ca 2ϩ -ATPase with TNP-ADP. This method can be used to determine whether or not mastoparan alters the affinity of the ATPase for TNP-ADP and hence ATP. Under conditions where the ATPase is in the E1 conformation, we determined that the apparent K d for TNP-ADP was 3.76 Ϯ 0.61 M (data not shown), consistent with TNP-ADP binding to the high affinity nucleotide binding site (28). In the presence of 70 M mastoparan, this value was relatively unchanged, 3.45 Ϯ 0.48 M.
As shown in Fig. 7, mastoparan causes a Ϸ40% reduction in the levels of phosphoenzyme formed from 1 mM [ 32 P]P i at 25°C. The presence of mastoparan therefore shifts the equilibrium between E2 and E2⅐P i toward E2.
Mastoparan Analogues-Since mastoparan was first discovered several other analogues have been isolated from different species of wasp. In addition several synthetic analogues have been made. The effect of a selection of these analogues on the ATPase was undertaken, and the results were compared with those obtained for mastoparan. Table III summarizes the effect of these peptides on the Ca 2ϩ -ATPase. The fractional ␣-helical contents and hydrophobic moments of the peptides derived from other studies are also shown in this table (15,29,30).
MP17 is a relatively inactive synthetic mastoparan analogue, in which two amino acid substitutions reduce ␣-helix formation as well as decrease the hydrophobic moment of the peptide (15). It can be seen that this peptide is a far weaker inhibitor of the ATPase. The K i value for MP17 is extrapolated, because at 30 M MP17 only 10% inhibition was observed. This peptide also has no effect on the fluorescence intensity of NBDlabeled ATPase.
MPX and MP7 retain the ability to form ␣-helices in lipids and have similar hydrophobic moments to mastoparan. They also inhibit the ATPase, with K i values of 4.4 and 2.2 M, respectively. Similar to mastoparan, these two analogues increase the fluorescence intensity of NBD-labeled ATPase, implying that they too shift the E2-E1 equilibrium toward E1.
Conclusion-A number of P 2 -type ATPases, including the Ca 2ϩ -ATPase, H ϩ /K ϩ -ATPase, and Na ϩ /K ϩ -ATPase are inhibited by peptide toxins such as mastoparan and melittin (4 -10, 13). Because this family of enzymes are believed to operate by similar mechanisms, a detailed understanding of how mastoparan inhibits the Ca 2ϩ -ATPase may also provide insight into the mechanism of inhibition of other P 2 -type ATPases by this peptide.
Mastoparan acts as a potent inhibitor of the SR Ca 2ϩ -ATPase, having a K i value of 1 Ϯ 0.13 M. The two analogues, MPX and MP7, also inhibit the ATPase with inhibitory constants of between 2.2 and 4.4 M. Thus mastoparan acts as one of the most potent ATPase inhibitors, apart from thapsigargin (31).
Using both NBD-labeled ATPase and FITC-labeled ATPase, mastoparan has been shown to shift the E2-E1 equilibrium of the enzyme toward E1, in contrast to thapsigargin which stabilizes the E2 form of the enzyme (31). At pH 6.0, the maximum percent increase in NBD fluorescence is approximately twice that at pH 7.2, because more of the ATPase can be shifted from the E2 form to the E1 form. Froud and Lee (23) have shown that the ratio E1/E2 varies from 0.1 to 0.5 with changing pH from 6 to 7. The apparent affinity of mastoparan for the ATPase has been reduced at pH 6.0, demonstrating that ionic interactions are important in the binding of the peptide to the ATPase.
Mastoparan may also affect the regulatory ATP binding site, because the V max for the regulatory site is reduced from 4.34 IU/mg to 1.24 IU/mg in the presence of the inhibitor.
The presence of mastoparan halves the maximum change in tryptophan fluorescence intensity observed, suggesting that mastoparan has altered the normal sequence of conformational changes that occur either on Ca 2ϩ binding or Ca 2ϩ dissociation.
The Ca 2ϩ dependence of ATPase activity data suggested that the affinity of Ca 2ϩ binding to the E1 form of the ATPase was reduced by Ϸ2-fold in the presence of mastoparan. As a result, the Ca 2ϩ binding step was isolated and investigated in more detail. Both the tryptophan fluorescence data and the Ca 2ϩ binding studies suggested that mastoparan decreased the affinity of the ATPase for Ca 2ϩ by between Ϸ10and 20-fold. The stoichiometry of Ca 2ϩ binding, however, was shown to be unaffected.
In the absence of mastoparan, binding of Ca 2ϩ to the ATPase is cooperative, whereas this cooperativity is abolished in the presence of mastoparan. This suggests that the Ca 2ϩ binding sites have been effected in such a way that Ca 2ϩ binding is no longer a two-step process, comprising fast binding of a first Ca 2ϩ ion followed by a slow conformational change, which allows binding of a second Ca 2ϩ ion. In the presence of mastoparan, the peptide appears to have altered the Ca 2ϩ binding sites in such a way that the two Ca 2ϩ ions now bind independently of one another. Thus, binding of the second Ca 2ϩ ion is not dependent on a conformational change produced by binding of the first Ca 2ϩ ion.
Site-directed mutagenesis has localized the two calcium binding sites between transmembrane helices M4, M5, and M6 (32,33). The critical residues have been identified as Glu 309 , Glu 771 , Asp 800 , Thr 799 , and Asn 796 . In addition Glu 908 , located within transmembrane helix M8, may play a minimal role in Ca 2ϩ binding. Two models have been proposed to describe the relationship between these two sites; the "stacked" model and the more recently proposed "side-by-side" model (34). Although both models can account for the cooperative nature of Ca 2ϩ binding, the later model also allows the possibility of independent Ca 2ϩ entry to the two sites. Our results appear to support the side-by-side model, because the presence of mastoparan causes a conformational change that results in independent Ca 2ϩ binding (see Scheme 1).
The data from the rapid kinetic measurements clearly show that in the absence of mastoparan both Ca 2ϩ binding and dissociation to and from the ATPase are biphasic in nature. In the presence of mastoparan these processes are now monophasic, lending further weight to the argument that Ca 2ϩ binding/  dissociation is independent in the presence of the peptide. Mg 2ϩ is an essential activator of the SR Ca 2ϩ -ATPase. It is required for several of the steps, which together make up the catalytic cycle of the enzyme. However, Mg 2ϩ is also in competition with Ca 2ϩ for binding at the two Ca 2ϩ binding sites. Consequently, ATPase activity decreases with increasing concentrations of Mg 2ϩ in the absence of mastoparan. At low concentrations of Mg 2ϩ , the presence of mastoparan results in strong inhibition of the ATPase. As the Mg 2ϩ concentration is increased from 2 to 10 mM, stimulation of ATPase activity is then observed, followed by inhibition. Perhaps the stimulatory effects of Mg 2ϩ (which are usually masked by the competition between Ca 2ϩ and Mg 2ϩ ) are due to the effect of mastoparan on the Ca 2ϩ binding sites. We have determined that the affinity of the Ca 2ϩ binding sites for Ca 2ϩ is reduced in the presence of mastoparan, it could therefore follow that the affinity of these sites for Mg 2ϩ will also be reduced. For the stimulatory effects of Mg 2ϩ to be observed, the affinity of the Ca 2ϩ binding sites for Mg 2ϩ must have been more greatly reduced than they were for Ca 2ϩ .
The effect of mastoparan on the steady-state levels of phosphorylation of the ATPase by [␥-32 P]ATP was studied. The results suggest that mastoparan increases the affinity of the ATPase for ATP and that the steady-state equilibrium between E1ЈCa 2 and E1ЈCa 2 ⅐P is pushed toward E1ЈCa 2 ⅐P. However, mastoparan was shown not to effect the binding of ATP to the catalytic site (see Fig. 1C) and was also shown to have no effect on the affinity of the enzyme for TNP-ADP. Petithory and Jencks (35) have suggested that phosphorylation of the ATPase by ATP is a two-step process in which ATP binding is followed by a conformational change. This active conformation of the enzyme is then able to undergo rapid phosphorylation.
Because binding of ATP is unaffected by mastoparan, perhaps the peptide effects the conformational change associated with ATP binding, thus decreasing the apparent K d for ATP in the phosphorylation experiment.
Mastoparan causes a reduction in the levels of phosphoenzyme formed from 1 mM [ 32 P]P i , thereby pushing the E2 7 E2⅐P i equilibrium toward E2. The reduction in levels of phosphoenzyme could be due to reduced levels of E2 as a result of mastoparan shifting the E2-E1 transition toward E1.
The two mastoparan analogues, MPX and MP7, have similar effects on ATPase activity as mastoparan itself. MP17, however does not. MP17 has a reduced ␣-helical content in membranes and a much smaller hydrophobic moment. Thus, the ability of mastoparan analogues to interact with and inhibit the Ca 2ϩ -ATPase appears to correlate with their ability to adopt ordered conformations in membranes as well as their amphiphilicity. SCHEME 1. A and B represent the stacked and side-by-side models that have been proposed to describe the relationship between the two Ca 2ϩ binding sites (I and II). Large arrows indicate the major route of Ca 2ϩ translocation, whereas the smaller arrows indicate possible other sites of Ca 2ϩ entry and exit. C represents the two sites in the presence of mastoparan. Ca 2ϩ binding is now seen to be independent.

TABLE III
Comparison of the effects of mastoparan analogues on the ATPase with mastoparan corresponds to the hydrophobic moment of the peptide (a measurement of the asymmetry with which hydrophobicity is distributed around the axis of a helix). Fractional ␣-helical contents and hydrophobic moments are derived from other studies (see Refs. 15,29,and 30